# Scale Labs SEAL Japanese (SEAL Japanese)

> Scale's SEAL leaderboard for Japanese: expert raters judge how well models answer about 1,000 prompts written by native Japanese speakers. BenchLM shows it as display-only reference data, excluded from rankings. Scale marked this board deprecated as of March 2025 and no longer adds models.

Canonical page: https://benchlm.ai/benchmarks/scale-japanese

- Category: [external](/external)
- Last updated: November 12, 2026 snapshot

## About SEAL Japanese

- Year: 2026
- Tasks: About 1,000 prompts per language
- Format: Elo-style rating
- Difficulty: External agent and model evaluation
- Paper: [Scale Labs leaderboard](https://labs.scale.com/leaderboard/japanese)

Single-turn prompts across general and culturally specific scenarios, rated by native-speaking experts. Scores are ratings on an Elo-style scale built from expert human comparisons, so only the gaps between rows carry meaning. BenchLM mirrors 17 published rows captured on November 12, 2026 snapshot.

SEAL Japanese is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (17 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [o1-preview](/models/o1-preview) | OpenAI | 1118 |
| 2 | [Gemini 2.0 Pro (December 2024)](https://labs.scale.com/leaderboard/japanese) | Google | 1117 |
| 3 | [o3-mini](/models/o3-mini) | OpenAI | 1104 |
| 4 | [Gemini 1.5 Pro (August 27, 2024)](https://labs.scale.com/leaderboard/japanese) | Google | 1093 |
| 5 | [o1 (December 2024)](https://labs.scale.com/leaderboard/japanese) | OpenAI | 1086 |
| 6 | [Gemini 1.5 Pro (November 2024)](https://labs.scale.com/leaderboard/japanese) | Google | 1084 |
| 7 | [GPT-4o (August 2024)](https://labs.scale.com/leaderboard/japanese) | OpenAI | 1077 |
| 8 | [Claude 3.5 Sonnet (June 2024)](https://labs.scale.com/leaderboard/japanese) | Anthropic | 1069 |
| 9 | [Gemini Pro Flash 2](https://labs.scale.com/leaderboard/japanese) | Google | 1064 |
| 10 | [GPT-4 (November 2024)](https://labs.scale.com/leaderboard/japanese) | OpenAI | 978 |
| 11 | [Mistral Large 2](/models/mistral-large-2) | Mistral | 967 |
| 12 | [Gemini 1.5 Flash](/models/gemini-1-5-flash) | Google | 966 |
| 13 | [Aya Expanse 32B](https://labs.scale.com/leaderboard/japanese) | Cohere | 960 |
| 14 | [Llama 3.1 405B Instruct](https://labs.scale.com/leaderboard/japanese) | Meta | 895 |
| 15 | [Gemma 2 27B](/models/gemma-2-27b) | Google | 892 |
| 16 | [Llama 3.3 70B Instruct](https://labs.scale.com/leaderboard/japanese) | Meta | 778 |
| 17 | [Aya 23 35B*](https://labs.scale.com/leaderboard/japanese) | Cohere | 760 |

## FAQ

### What does SEAL Japanese measure?

Scale's SEAL leaderboard for Japanese: expert raters judge how well models answer about 1,000 prompts written by native Japanese speakers. BenchLM shows it as display-only reference data, excluded from rankings. Scale marked this board deprecated as of March 2025 and no longer adds models.

### Which model leads the published SEAL Japanese snapshot?

o1-preview currently leads the published SEAL Japanese snapshot with a score of 1118.

### How many models are evaluated on SEAL Japanese?

The November 12, 2026 snapshot contains 17 AI models.

### Does SEAL Japanese affect BenchLM's overall score?

Not directly. SEAL Japanese is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.